Properties of Counting Measures Restricted to a Particular Sigma Algebra
up vote
2
down vote
favorite
Define $mathcal{F}={Asubseteqmathbb{R} vert 0in A^{mathrm{o}} or 0 in (A^c)^{mathrm{o}} }$, where $A^{mathrm{o}}$ is the interior of $A$. It can be shown quite easily that $mathcal{F}$ is an algebra of sets, and that $sigma(mathcal{F})$ contains the singletons of $mathbb{R}$.
If $gamma$ is the counting measure on $(mathbb{R},mathcal{P}(mathbb{R}))$, let $alpha(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} })$ and $beta(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} } cup {0})$. If we define $mu := left.alpha right|_{sigma(mathcal{F})}$ and $nu := left.beta right|_{sigma(mathcal{F})}$ (the restrictions of $mu$ and $nu$ to
$sigma(mathcal{F})$) then it turns out that:
a) both $mu$ and $nu$ are $sigma$-finite.
b) $mu$ and $nu$ coincide on $mathcal{F}$ (that is, $mu(A)=nu(A)$, $forall A in mathcal{F}$).
c) $mu neq nu(A)$ on $sigma(mathcal{F})$.
I am having trouble in proving these 3 facts, and any help/hints to help push me towards the right direction will be much appreciated.
In particular, I'm having difficulty in working with any general set $Ain sigma(mathcal{F})$; I currently have no clue on how to construct a sequence $(A_n)_{n=1}^{infty}$ in $sigma(mathcal{F})$ such that $bigcuplimits_{n=1}^{infty}A_n = mathbb{R}$ with $mu(A_n)<infty$ and $nu(A_n)<infty$.
EDIT: I've actually made some progress with this since asking: I have solutions to b) and c), and have partial work for a). The struggle in part a) remains with constructing such a sequence of sets. Best I've come up with is $(A_k)_{k=1}^{infty}=(-k,-frac{1}{k}] cup [frac{1}{k},k)$, and in this case, $0 in (A_k^c)^{mathrm{o}}$ with $mu(A_k)<infty$ and $nu(A_k)<infty$. Problem here is that the countable union isn't the whole real line (it misses ${0}$!). Similarly, I've also considered the sequence $(A_k)_{k=1}^{infty}=(-k,0] cup [frac{1}{k},k)$, whose countable union is all of $mathbb{R}$, $mu$ and $nu$ are finite $forall k$, but this sequence is not in $mathcal{F}$.
real-analysis analysis measure-theory
This question has an open bounty worth +50
reputation from Jane Doe ending in 6 days.
This question has not received enough attention.
Looking for a complete answer or detailed hints with a suggest source for further information.
add a comment |
up vote
2
down vote
favorite
Define $mathcal{F}={Asubseteqmathbb{R} vert 0in A^{mathrm{o}} or 0 in (A^c)^{mathrm{o}} }$, where $A^{mathrm{o}}$ is the interior of $A$. It can be shown quite easily that $mathcal{F}$ is an algebra of sets, and that $sigma(mathcal{F})$ contains the singletons of $mathbb{R}$.
If $gamma$ is the counting measure on $(mathbb{R},mathcal{P}(mathbb{R}))$, let $alpha(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} })$ and $beta(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} } cup {0})$. If we define $mu := left.alpha right|_{sigma(mathcal{F})}$ and $nu := left.beta right|_{sigma(mathcal{F})}$ (the restrictions of $mu$ and $nu$ to
$sigma(mathcal{F})$) then it turns out that:
a) both $mu$ and $nu$ are $sigma$-finite.
b) $mu$ and $nu$ coincide on $mathcal{F}$ (that is, $mu(A)=nu(A)$, $forall A in mathcal{F}$).
c) $mu neq nu(A)$ on $sigma(mathcal{F})$.
I am having trouble in proving these 3 facts, and any help/hints to help push me towards the right direction will be much appreciated.
In particular, I'm having difficulty in working with any general set $Ain sigma(mathcal{F})$; I currently have no clue on how to construct a sequence $(A_n)_{n=1}^{infty}$ in $sigma(mathcal{F})$ such that $bigcuplimits_{n=1}^{infty}A_n = mathbb{R}$ with $mu(A_n)<infty$ and $nu(A_n)<infty$.
EDIT: I've actually made some progress with this since asking: I have solutions to b) and c), and have partial work for a). The struggle in part a) remains with constructing such a sequence of sets. Best I've come up with is $(A_k)_{k=1}^{infty}=(-k,-frac{1}{k}] cup [frac{1}{k},k)$, and in this case, $0 in (A_k^c)^{mathrm{o}}$ with $mu(A_k)<infty$ and $nu(A_k)<infty$. Problem here is that the countable union isn't the whole real line (it misses ${0}$!). Similarly, I've also considered the sequence $(A_k)_{k=1}^{infty}=(-k,0] cup [frac{1}{k},k)$, whose countable union is all of $mathbb{R}$, $mu$ and $nu$ are finite $forall k$, but this sequence is not in $mathcal{F}$.
real-analysis analysis measure-theory
This question has an open bounty worth +50
reputation from Jane Doe ending in 6 days.
This question has not received enough attention.
Looking for a complete answer or detailed hints with a suggest source for further information.
you don't have to construct a sequence which the union is equal $sigma(mathcal{F})$, but $X$ instead. As $Xin sigma(mathcal{F})$ the sigma-algebra is a family of sets, not a kind of 'measure subspace'
– Robson
Nov 18 at 22:33
Thank you, I'll make the edit!
– Buttfor
Nov 18 at 22:39
Why do you want $A_k$ to be in $mathcal{F}$? For sigma-finite you only need $A_kin sigma(mathcal{F})$.
– Severin Schraven
yesterday
In fact, your proposed $A_k = (-k, 0] cup [frac{1}{k}, k)$ works just fine.
– Severin Schraven
yesterday
add a comment |
up vote
2
down vote
favorite
up vote
2
down vote
favorite
Define $mathcal{F}={Asubseteqmathbb{R} vert 0in A^{mathrm{o}} or 0 in (A^c)^{mathrm{o}} }$, where $A^{mathrm{o}}$ is the interior of $A$. It can be shown quite easily that $mathcal{F}$ is an algebra of sets, and that $sigma(mathcal{F})$ contains the singletons of $mathbb{R}$.
If $gamma$ is the counting measure on $(mathbb{R},mathcal{P}(mathbb{R}))$, let $alpha(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} })$ and $beta(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} } cup {0})$. If we define $mu := left.alpha right|_{sigma(mathcal{F})}$ and $nu := left.beta right|_{sigma(mathcal{F})}$ (the restrictions of $mu$ and $nu$ to
$sigma(mathcal{F})$) then it turns out that:
a) both $mu$ and $nu$ are $sigma$-finite.
b) $mu$ and $nu$ coincide on $mathcal{F}$ (that is, $mu(A)=nu(A)$, $forall A in mathcal{F}$).
c) $mu neq nu(A)$ on $sigma(mathcal{F})$.
I am having trouble in proving these 3 facts, and any help/hints to help push me towards the right direction will be much appreciated.
In particular, I'm having difficulty in working with any general set $Ain sigma(mathcal{F})$; I currently have no clue on how to construct a sequence $(A_n)_{n=1}^{infty}$ in $sigma(mathcal{F})$ such that $bigcuplimits_{n=1}^{infty}A_n = mathbb{R}$ with $mu(A_n)<infty$ and $nu(A_n)<infty$.
EDIT: I've actually made some progress with this since asking: I have solutions to b) and c), and have partial work for a). The struggle in part a) remains with constructing such a sequence of sets. Best I've come up with is $(A_k)_{k=1}^{infty}=(-k,-frac{1}{k}] cup [frac{1}{k},k)$, and in this case, $0 in (A_k^c)^{mathrm{o}}$ with $mu(A_k)<infty$ and $nu(A_k)<infty$. Problem here is that the countable union isn't the whole real line (it misses ${0}$!). Similarly, I've also considered the sequence $(A_k)_{k=1}^{infty}=(-k,0] cup [frac{1}{k},k)$, whose countable union is all of $mathbb{R}$, $mu$ and $nu$ are finite $forall k$, but this sequence is not in $mathcal{F}$.
real-analysis analysis measure-theory
Define $mathcal{F}={Asubseteqmathbb{R} vert 0in A^{mathrm{o}} or 0 in (A^c)^{mathrm{o}} }$, where $A^{mathrm{o}}$ is the interior of $A$. It can be shown quite easily that $mathcal{F}$ is an algebra of sets, and that $sigma(mathcal{F})$ contains the singletons of $mathbb{R}$.
If $gamma$ is the counting measure on $(mathbb{R},mathcal{P}(mathbb{R}))$, let $alpha(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} })$ and $beta(A)=gamma(A cap {frac{1}{n} vert n in mathbb{N} } cup {0})$. If we define $mu := left.alpha right|_{sigma(mathcal{F})}$ and $nu := left.beta right|_{sigma(mathcal{F})}$ (the restrictions of $mu$ and $nu$ to
$sigma(mathcal{F})$) then it turns out that:
a) both $mu$ and $nu$ are $sigma$-finite.
b) $mu$ and $nu$ coincide on $mathcal{F}$ (that is, $mu(A)=nu(A)$, $forall A in mathcal{F}$).
c) $mu neq nu(A)$ on $sigma(mathcal{F})$.
I am having trouble in proving these 3 facts, and any help/hints to help push me towards the right direction will be much appreciated.
In particular, I'm having difficulty in working with any general set $Ain sigma(mathcal{F})$; I currently have no clue on how to construct a sequence $(A_n)_{n=1}^{infty}$ in $sigma(mathcal{F})$ such that $bigcuplimits_{n=1}^{infty}A_n = mathbb{R}$ with $mu(A_n)<infty$ and $nu(A_n)<infty$.
EDIT: I've actually made some progress with this since asking: I have solutions to b) and c), and have partial work for a). The struggle in part a) remains with constructing such a sequence of sets. Best I've come up with is $(A_k)_{k=1}^{infty}=(-k,-frac{1}{k}] cup [frac{1}{k},k)$, and in this case, $0 in (A_k^c)^{mathrm{o}}$ with $mu(A_k)<infty$ and $nu(A_k)<infty$. Problem here is that the countable union isn't the whole real line (it misses ${0}$!). Similarly, I've also considered the sequence $(A_k)_{k=1}^{infty}=(-k,0] cup [frac{1}{k},k)$, whose countable union is all of $mathbb{R}$, $mu$ and $nu$ are finite $forall k$, but this sequence is not in $mathcal{F}$.
real-analysis analysis measure-theory
real-analysis analysis measure-theory
edited Nov 21 at 2:56
asked Nov 18 at 19:33
Buttfor
294
294
This question has an open bounty worth +50
reputation from Jane Doe ending in 6 days.
This question has not received enough attention.
Looking for a complete answer or detailed hints with a suggest source for further information.
This question has an open bounty worth +50
reputation from Jane Doe ending in 6 days.
This question has not received enough attention.
Looking for a complete answer or detailed hints with a suggest source for further information.
you don't have to construct a sequence which the union is equal $sigma(mathcal{F})$, but $X$ instead. As $Xin sigma(mathcal{F})$ the sigma-algebra is a family of sets, not a kind of 'measure subspace'
– Robson
Nov 18 at 22:33
Thank you, I'll make the edit!
– Buttfor
Nov 18 at 22:39
Why do you want $A_k$ to be in $mathcal{F}$? For sigma-finite you only need $A_kin sigma(mathcal{F})$.
– Severin Schraven
yesterday
In fact, your proposed $A_k = (-k, 0] cup [frac{1}{k}, k)$ works just fine.
– Severin Schraven
yesterday
add a comment |
you don't have to construct a sequence which the union is equal $sigma(mathcal{F})$, but $X$ instead. As $Xin sigma(mathcal{F})$ the sigma-algebra is a family of sets, not a kind of 'measure subspace'
– Robson
Nov 18 at 22:33
Thank you, I'll make the edit!
– Buttfor
Nov 18 at 22:39
Why do you want $A_k$ to be in $mathcal{F}$? For sigma-finite you only need $A_kin sigma(mathcal{F})$.
– Severin Schraven
yesterday
In fact, your proposed $A_k = (-k, 0] cup [frac{1}{k}, k)$ works just fine.
– Severin Schraven
yesterday
you don't have to construct a sequence which the union is equal $sigma(mathcal{F})$, but $X$ instead. As $Xin sigma(mathcal{F})$ the sigma-algebra is a family of sets, not a kind of 'measure subspace'
– Robson
Nov 18 at 22:33
you don't have to construct a sequence which the union is equal $sigma(mathcal{F})$, but $X$ instead. As $Xin sigma(mathcal{F})$ the sigma-algebra is a family of sets, not a kind of 'measure subspace'
– Robson
Nov 18 at 22:33
Thank you, I'll make the edit!
– Buttfor
Nov 18 at 22:39
Thank you, I'll make the edit!
– Buttfor
Nov 18 at 22:39
Why do you want $A_k$ to be in $mathcal{F}$? For sigma-finite you only need $A_kin sigma(mathcal{F})$.
– Severin Schraven
yesterday
Why do you want $A_k$ to be in $mathcal{F}$? For sigma-finite you only need $A_kin sigma(mathcal{F})$.
– Severin Schraven
yesterday
In fact, your proposed $A_k = (-k, 0] cup [frac{1}{k}, k)$ works just fine.
– Severin Schraven
yesterday
In fact, your proposed $A_k = (-k, 0] cup [frac{1}{k}, k)$ works just fine.
– Severin Schraven
yesterday
add a comment |
1 Answer
1
active
oldest
votes
up vote
1
down vote
1.) We have
$$ (-infty, 0] = bigcap_{ngeq 1} (-infty, frac{1}{n}) in sigma(mathcal{F}) $$
as well as $ [frac{1}{k}, k) $ for all $kin mathbb{N}$. We define
$$ A_k := (-infty, 0] cup [frac{1}{k}, k) in sigma(mathcal{F}).$$
We note that
$$ mathbb{R} = bigcup_{kgeq 1} A_k.$$
If we can show that those $A_k$ have finite measure for both $mu$ and $nu$, then we have shown that $mu$ and $nu$ are sigma-finite. Indeed, we have
$$ mu(A_k) = gamma ({ nin mathbb{N} : frac{1}{n} in A_k }) = gamma({1, dots, k}) = k. $$
Similarly, we have
$$ nu(A_k) = k+1 $$
(the plus one comes from $0in A_k$). We conclude that $mu, nu$ are sigma-finite.
2.) Pick $Ain mathcal{F}$. By the very definition of $mathcal{F}$ we have two cases:
Firstly, we consider the case $0in A^o$. This means, that there exists $1>varepsilon>0$ such that $(-varepsilon, varepsilon)subseteq A$. However, then we have
$$ mu(A) = infty = nu(A) $$
as there exists $N>1$ such that for all $ngeq N$ we have $0<frac{1}{n}<varepsilon <frac{1}{N}$ and hence
$$ mu(A) geq mu((-varepsilon, varepsilon)) = gamma({ nin mathbb{N} : n>N })= infty. $$
Similarly one shows that $nu(A)=infty$.
In the second case we have that $0in (mathbb{R}setminus A)^o$. Then we have that $0notin A$ and thus, we have $mu(A)=nu(A)$ as we have by definition
$$ nu(A) = mu(A) + gamma(Acap {0}). $$
c.) For this we recall that the singelons are in the sigma-algebra and hence, we have ${0}in sigma(mathcal{F})$ and
$$ nu({0})= 1 neq 0 = mu({0}).$$
add a comment |
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
1
down vote
1.) We have
$$ (-infty, 0] = bigcap_{ngeq 1} (-infty, frac{1}{n}) in sigma(mathcal{F}) $$
as well as $ [frac{1}{k}, k) $ for all $kin mathbb{N}$. We define
$$ A_k := (-infty, 0] cup [frac{1}{k}, k) in sigma(mathcal{F}).$$
We note that
$$ mathbb{R} = bigcup_{kgeq 1} A_k.$$
If we can show that those $A_k$ have finite measure for both $mu$ and $nu$, then we have shown that $mu$ and $nu$ are sigma-finite. Indeed, we have
$$ mu(A_k) = gamma ({ nin mathbb{N} : frac{1}{n} in A_k }) = gamma({1, dots, k}) = k. $$
Similarly, we have
$$ nu(A_k) = k+1 $$
(the plus one comes from $0in A_k$). We conclude that $mu, nu$ are sigma-finite.
2.) Pick $Ain mathcal{F}$. By the very definition of $mathcal{F}$ we have two cases:
Firstly, we consider the case $0in A^o$. This means, that there exists $1>varepsilon>0$ such that $(-varepsilon, varepsilon)subseteq A$. However, then we have
$$ mu(A) = infty = nu(A) $$
as there exists $N>1$ such that for all $ngeq N$ we have $0<frac{1}{n}<varepsilon <frac{1}{N}$ and hence
$$ mu(A) geq mu((-varepsilon, varepsilon)) = gamma({ nin mathbb{N} : n>N })= infty. $$
Similarly one shows that $nu(A)=infty$.
In the second case we have that $0in (mathbb{R}setminus A)^o$. Then we have that $0notin A$ and thus, we have $mu(A)=nu(A)$ as we have by definition
$$ nu(A) = mu(A) + gamma(Acap {0}). $$
c.) For this we recall that the singelons are in the sigma-algebra and hence, we have ${0}in sigma(mathcal{F})$ and
$$ nu({0})= 1 neq 0 = mu({0}).$$
add a comment |
up vote
1
down vote
1.) We have
$$ (-infty, 0] = bigcap_{ngeq 1} (-infty, frac{1}{n}) in sigma(mathcal{F}) $$
as well as $ [frac{1}{k}, k) $ for all $kin mathbb{N}$. We define
$$ A_k := (-infty, 0] cup [frac{1}{k}, k) in sigma(mathcal{F}).$$
We note that
$$ mathbb{R} = bigcup_{kgeq 1} A_k.$$
If we can show that those $A_k$ have finite measure for both $mu$ and $nu$, then we have shown that $mu$ and $nu$ are sigma-finite. Indeed, we have
$$ mu(A_k) = gamma ({ nin mathbb{N} : frac{1}{n} in A_k }) = gamma({1, dots, k}) = k. $$
Similarly, we have
$$ nu(A_k) = k+1 $$
(the plus one comes from $0in A_k$). We conclude that $mu, nu$ are sigma-finite.
2.) Pick $Ain mathcal{F}$. By the very definition of $mathcal{F}$ we have two cases:
Firstly, we consider the case $0in A^o$. This means, that there exists $1>varepsilon>0$ such that $(-varepsilon, varepsilon)subseteq A$. However, then we have
$$ mu(A) = infty = nu(A) $$
as there exists $N>1$ such that for all $ngeq N$ we have $0<frac{1}{n}<varepsilon <frac{1}{N}$ and hence
$$ mu(A) geq mu((-varepsilon, varepsilon)) = gamma({ nin mathbb{N} : n>N })= infty. $$
Similarly one shows that $nu(A)=infty$.
In the second case we have that $0in (mathbb{R}setminus A)^o$. Then we have that $0notin A$ and thus, we have $mu(A)=nu(A)$ as we have by definition
$$ nu(A) = mu(A) + gamma(Acap {0}). $$
c.) For this we recall that the singelons are in the sigma-algebra and hence, we have ${0}in sigma(mathcal{F})$ and
$$ nu({0})= 1 neq 0 = mu({0}).$$
add a comment |
up vote
1
down vote
up vote
1
down vote
1.) We have
$$ (-infty, 0] = bigcap_{ngeq 1} (-infty, frac{1}{n}) in sigma(mathcal{F}) $$
as well as $ [frac{1}{k}, k) $ for all $kin mathbb{N}$. We define
$$ A_k := (-infty, 0] cup [frac{1}{k}, k) in sigma(mathcal{F}).$$
We note that
$$ mathbb{R} = bigcup_{kgeq 1} A_k.$$
If we can show that those $A_k$ have finite measure for both $mu$ and $nu$, then we have shown that $mu$ and $nu$ are sigma-finite. Indeed, we have
$$ mu(A_k) = gamma ({ nin mathbb{N} : frac{1}{n} in A_k }) = gamma({1, dots, k}) = k. $$
Similarly, we have
$$ nu(A_k) = k+1 $$
(the plus one comes from $0in A_k$). We conclude that $mu, nu$ are sigma-finite.
2.) Pick $Ain mathcal{F}$. By the very definition of $mathcal{F}$ we have two cases:
Firstly, we consider the case $0in A^o$. This means, that there exists $1>varepsilon>0$ such that $(-varepsilon, varepsilon)subseteq A$. However, then we have
$$ mu(A) = infty = nu(A) $$
as there exists $N>1$ such that for all $ngeq N$ we have $0<frac{1}{n}<varepsilon <frac{1}{N}$ and hence
$$ mu(A) geq mu((-varepsilon, varepsilon)) = gamma({ nin mathbb{N} : n>N })= infty. $$
Similarly one shows that $nu(A)=infty$.
In the second case we have that $0in (mathbb{R}setminus A)^o$. Then we have that $0notin A$ and thus, we have $mu(A)=nu(A)$ as we have by definition
$$ nu(A) = mu(A) + gamma(Acap {0}). $$
c.) For this we recall that the singelons are in the sigma-algebra and hence, we have ${0}in sigma(mathcal{F})$ and
$$ nu({0})= 1 neq 0 = mu({0}).$$
1.) We have
$$ (-infty, 0] = bigcap_{ngeq 1} (-infty, frac{1}{n}) in sigma(mathcal{F}) $$
as well as $ [frac{1}{k}, k) $ for all $kin mathbb{N}$. We define
$$ A_k := (-infty, 0] cup [frac{1}{k}, k) in sigma(mathcal{F}).$$
We note that
$$ mathbb{R} = bigcup_{kgeq 1} A_k.$$
If we can show that those $A_k$ have finite measure for both $mu$ and $nu$, then we have shown that $mu$ and $nu$ are sigma-finite. Indeed, we have
$$ mu(A_k) = gamma ({ nin mathbb{N} : frac{1}{n} in A_k }) = gamma({1, dots, k}) = k. $$
Similarly, we have
$$ nu(A_k) = k+1 $$
(the plus one comes from $0in A_k$). We conclude that $mu, nu$ are sigma-finite.
2.) Pick $Ain mathcal{F}$. By the very definition of $mathcal{F}$ we have two cases:
Firstly, we consider the case $0in A^o$. This means, that there exists $1>varepsilon>0$ such that $(-varepsilon, varepsilon)subseteq A$. However, then we have
$$ mu(A) = infty = nu(A) $$
as there exists $N>1$ such that for all $ngeq N$ we have $0<frac{1}{n}<varepsilon <frac{1}{N}$ and hence
$$ mu(A) geq mu((-varepsilon, varepsilon)) = gamma({ nin mathbb{N} : n>N })= infty. $$
Similarly one shows that $nu(A)=infty$.
In the second case we have that $0in (mathbb{R}setminus A)^o$. Then we have that $0notin A$ and thus, we have $mu(A)=nu(A)$ as we have by definition
$$ nu(A) = mu(A) + gamma(Acap {0}). $$
c.) For this we recall that the singelons are in the sigma-algebra and hence, we have ${0}in sigma(mathcal{F})$ and
$$ nu({0})= 1 neq 0 = mu({0}).$$
answered yesterday
Severin Schraven
5,4511832
5,4511832
add a comment |
add a comment |
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3004018%2fproperties-of-counting-measures-restricted-to-a-particular-sigma-algebra%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
you don't have to construct a sequence which the union is equal $sigma(mathcal{F})$, but $X$ instead. As $Xin sigma(mathcal{F})$ the sigma-algebra is a family of sets, not a kind of 'measure subspace'
– Robson
Nov 18 at 22:33
Thank you, I'll make the edit!
– Buttfor
Nov 18 at 22:39
Why do you want $A_k$ to be in $mathcal{F}$? For sigma-finite you only need $A_kin sigma(mathcal{F})$.
– Severin Schraven
yesterday
In fact, your proposed $A_k = (-k, 0] cup [frac{1}{k}, k)$ works just fine.
– Severin Schraven
yesterday